AI incident #571 · 22 June 2023

Reported Accidental Exposure of 38TB of Data by Microsoft's AI Research Team

Open on the AI Incident Database 1 news report Snapshot 7 Sep 2026

What happened

Microsoft's AI research team reportedly accidentally exposed 38TB of sensitive data while publishing open-source training material on GitHub. The exposure allegedly included secrets, private keys, passwords, and internal Microsoft Teams messages. The team reportedly utilized Azure's Shared Access Signature (SAS) tokens for sharing, which were purportedly misconfigured, leading to the wide exposure of data.

Only the incident metadata is stored here. The underlying news reports are on the AI Incident Database (CC BY-SA 4.0); use the links above to read them.

News reports (1)

Coverage catalogued by the AI Incident Database. Titles link to the original publisher; the text is not reproduced here.

18 Sep 2023
38TB of data accidentally exposed by Microsoft AI researchers
wiz.io · Hillai Ben-Sasson, Ronny Greenberg` · AIID #3221

Who was involved

Alleged deployer: Microsoft
Alleged developer: Microsoft'S Ai Research Division
Alleged harmed party: Microsoft Employees, Microsoft, Privacy, Third Parties Whose Confidential Data Was Exposed

Classification (MIT AI Risk Repository taxonomy)

Risk domain: Privacy & Security
Risk subdomain: 2.1 Compromise of privacy by obtaining, leaking or correctly inferring sensitive information
Causal entity: Human
Intent: Unintentional
Timing: Post-deployment
Harm level: —
Sectors: —
Countries: —

Risk entries describing this failure mode

Entries from the MIT AI Risk Repository coded to subdomain 2.1.

Risks to privacy
"General- purpose AI models or systems can ‘leak’ information about individuals whose data was used in training. For future models trained on sensitive personal data like health or financial data, this may lead to partic...
International Scientific Report on the Safety of Advanced AI (Bengio2024)
Risks to privacy
"General- purpose AI systems can cause or contribute to violations of user privacy. Violations can occur inadvertently during the training or usage of AI systems, for example through unauthorised processing of personal d...
International AI Safety Report 2025 (Bengio2025)
Privacy Leakage
"Privacy Leakage means the generated content includes sensitive personal information"
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Private Training Data
"As recent LLMs continue to incorporate licensed, created, and publicly available data sources in their corpora, the potential to mix private data in the training corpora is significantly increased. The misused private d...
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Memorization in LLMs
"Memorization in LLMs refers to the capability to recover the training data with contextual prefixes. According to [88]–[90], given a PII entity x, which is memorized by a model F. Using a prompt p could force the model...
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Association in LLMs
"Association in LLMs refers to the capability to associate various pieces of information related to a person. According to [68], [86], given a pair of PII entities (xi , xj ), which is associated by a model F. Using a pr...
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Privacy Leakage
"The model is trained with personal data in the corpus and unintentionally exposing them during the conversation."
Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems (Cui2024)
Privacy and regulation violations
"Some of the broken systems discussed above are also very invasive of people’s privacy, controlling, for instance, the length of someone’s last romantic relationship [51]. More recently, ChatGPT was banned in Italy over...
Navigating the Landscape of AI Ethics and Responsibility (Cunha2023)

Incidents in the same risk subdomain

All incidents in this subdomain